文摘
Chemometric data analysis tools were applied to chromatographic data to identify the presenceof solvents in gasoline samples from gas stations in Minas Gerais state, Brazil. A training set of75 samples was formulated by mixing pure gasoline with various concentrations of four complexsolvents. The samples were analyzed by GC-MS, and the selected peaks were used in chemometricstudies. Hierarchical cluster analysis, HCA, was used to search for sample distribution patternsaccording to the solvent added. K-nearest neighbor (KNN) was used to create a classificationscheme to differentiate pure and mixed samples and to indicate the type of solvent present. HCArevealed a clear clustering tendency of samples containing the same solvent. However, only afterthe exclusion of lesser variables (peaks) by means of Fisher weights was it possible to separatesamples with low solvent concentrations. After optimization of the KNN algorithm, it was possibleto classify 88% of the samples of the training set correctly. To check the quality of the model,another group of samples was prepared with certified gasoline and the same solvents. Thealgorithm classified the great majority of the samples correctly once again.